WriteFreely on GKE Autopilot

WriteFreely is an open-source, minimalist, federated blogging platform written in Go — a lightweight Medium alternative for publishing clean, distraction-free writing. This module deploys WriteFreely on GKE Autopilot on top of the App_GKE foundation, which provisions and manages the shared Google Cloud and Kubernetes infrastructure.

This guide focuses on the cloud services WriteFreely uses and how to explore and operate them from the Google Cloud Console and the command line. For the mechanics common to every GKE application — Workload Identity, ingress, autoscaling, CI/CD, Cloud Armor, IAP, Binary Authorization, VPC Service Controls, backups, and the deployment lifecycle — refer to the App_GKE foundation guide rather than repeating them here.

---

What WriteFreely costs on RAD, and how that compares

Running WriteFreely on GKE Autopilot pods in your own Google Cloud project costs about $96.70/month in Google Cloud charges, on top of a one-off 175-credit RAD module fee charged once at deploy time. WriteFreely needs a MySQL database behind it, which together with the pod itself is most of that $96.70. In a RAD-managed project the module fee drops to 157.5 credits (10% lower) and the running cost is metered hourly in credits instead of billed directly by Google — see Pause it for free below for the one RAD-managed advantage a self-hosted VPS cannot match — or see the Cloud Run guide for the lower-cost option.

What you pay on RAD

	Own project	RAD-managed project
Module fee (one-off)	175 credits	157.5 credits (10% lower)
Build time	~3-6 credits per build, either way	~3-6 credits per build, either way
Google Cloud running cost	$96.70/month, billed by Google at list price	~93 credits/day, metered at list price plus RAD's margin

Default running cost in your own project

Resource	US$/month (Google list price, us-central1)
GKE Autopilot pod, 1x vCPU / 2 GiB	$39.68
Cloud SQL for MySQL, 1 vCPU / 3.75 GB, zonal, 10 GB SSD	$51.02
Cloud NAT and networking	$5.00
Cloud Storage (add-ons, backups)	$1.00
GKE cluster management fee	shared across every GKE app in the project; $0 if this is your only cluster, else ~73
Total	$96.70

The GKE cluster management fee is excluded from the total above because it is shared across every GKE app in the project — $0 if this is the only one, otherwise roughly $73/month split across however many you run.

How it compares

• write.as, run by the same team as WriteFreely, is the official hosted version of this software: its pricing page (checked 8 October 2026) lists a free tier, Pro at $6/mo (annual) or $9/mo (monthly) for up to three blogs with custom themes, and Team at $25/mo (annual) or $30/mo (monthly) for shared, collaborative publications.
• Pro is cheaper than RAD's $96.70/month own-project running cost because it's a shared multi-tenant service; RAD's module instead gives WriteFreely its own dedicated Cloud Run service and Cloud SQL database, which matters once you want full control of the data, uptime and federation identity rather than a shared host.
• A bare-VPS alternative also works: a Hetzner or DigitalOcean 2 vCPU/4GB box at about $24/mo, or a GCP Compute Engine e2-standard-2 at about $49/mo, is enough to self-host WriteFreely and MySQL together — cheaper in cash terms than RAD, with none of the managed backups, patching or monitoring RAD provides.

Pause it for free: delete a RAD-managed project, restore it when you need it

This is the headline advantage of a RAD-managed project, and it costs nothing while paused. Deleting a RAD-managed project unlinks its billing first, then asks Google to delete the project outright — Google's own 30-day recoverable soft delete. Unlike deleting one module, this does not tear down Cloud SQL, any VM, or the compute resource one by one: the whole project simply stops, and nothing is charged while it waits, because billing is already unlinked.

Restoring, within 30 days and only by the project's owner, asks Google to undelete the project and reattaches its billing account, then asks you to run Update on each deployment to confirm everything came back. Because nothing was individually destroyed, that Update finds the same resources already there — it is a check, not a rebuild, and an Update never charges the module fee again. This costs only a handful of credits (under US$1) in total for a typical 2-3-deployment chain.

What this needs: you must own the project (not one RAD only manages billing for), you must restore it yourself within 30 days — after that Google deletes it for good — and restoring is admitted like creating a new project, so your purchased credit balance must still clear the tier's floor (100 credits for the sandbox tier most study/demo use fits). Google says most services are fully working again within 36 hours of a restore.

One real gap: nightly backups are written to a bucket inside the project, and that bucket has Cloud Storage's soft-delete explicitly turned off, so it is very likely gone as soon as you delete the project — even though the project itself is recoverable for 30 days. If you've customised WriteFreely and want to keep that work, copy a backup out (to Google Drive, or a bucket outside the project) before deleting. For a default install with nothing irreplaceable in it, this does not matter.

Pay only while you use it, the other way: delete and redeploy

In your own project, or once 30 days have passed on a RAD-managed one, there is no free pause — deleting removes the resources for good, and bringing WriteFreely back means redeploying from scratch. That costs about 178-181 credits (the 175-credit module fee plus a build).

Deleting only saves money once WriteFreely would otherwise sit unused long enough to clear that redeploy cost against its own running cost — for WriteFreely, that's about 6 days or more. Most of this running cost is usually the database — it stops only once nothing else in the project uses it, so deleting this app while something else shares the project saves only its own compute share.

Keep data first: nightly backups go to a bucket inside the deployment and are deleted with it, so copy the latest backup out before deleting if you want to keep it.

Lab sessions and Managed Environments

Lab sessions, for training: a trainer runs a session for a class. Each participant gets the app in their own Google Cloud project for 15 minutes to 24 hours, within an allowance the trainer sets. Either the trainer funds every place, or each participant pays for their own. Everything is deleted when the session ends and unused credits go back to the trainer.

Managed Environments, for consultancies: a partner runs the app for a client from a ring-fenced wallet it funds, and settles with the client directly. If the wallet runs low, billing pauses and the data is kept, so nobody receives an unexpected charge. At the end the partner hands the project over and the deployments become the client's own.

Credits can be bought in more than 20 currencies, including XAF, XOF, NGN, GHS, KES and ZAR, by card, bank transfer or mobile money.

Sources (8 October 2026): Google Cloud Billing Catalog API list prices; radmodules.dev/pricing for RAD's own fees and daily-credit estimates; Google's project delete/restore documentation for the pause/restore claims. write.as plan prices from write.as/pricing, checked 8 October 2026.

1. Overview

WriteFreely runs as a single Go web workload on GKE Autopilot. The deployment wires together a focused set of Google Cloud services:

Capability	Google Cloud service	Notes
Compute	GKE Autopilot	Go pods, 1 vCPU / 2 GiB by default, horizontally autoscaled
Database	Cloud SQL for MySQL 8.0	Required — reached through the Cloud SQL Auth Proxy sidecar on 127.0.0.1
Object storage	Cloud Storage	A dedicated writefreely-uploads data bucket provisioned automatically
Secrets	Secret Manager	Three auto-generated AES-256 key secrets; database password
Ingress	Cloud Load Balancing	External LoadBalancer, optional custom domain + managed certificate

Sensible defaults worth knowing up front:

• MySQL 8.0 is the fixed engine. The shared application layer standardises on Cloud SQL for MySQL; the GKE variant leaves database_type unset and inherits it.
• Cloud SQL is reached via the Auth Proxy sidecar. On GKE enable_cloudsql_volume = true and the variant overrides DB_HOST = 127.0.0.1 — WriteFreely dials the loopback address where the cloud-sql-proxy sidecar listens on port 3306.
• The three AES-256 keys are generated automatically and stored in Secret Manager (cookies-auth, cookies-enc, email-key). They must never be rotated after first boot — rotating them logs out every user and makes previously encrypted email data undecryptable.
• A custom image is built, not pulled prebuilt. container_image_source = custom: the thin config-gen wrapper (renders config.ini, seeds the keys, runs writefreely db init) is built by Cloud Build and pushed to Artifact Registry.
• Minimum 1 replica is maintained (min_instance_count = 1, max_instance_count = 1; GKE does not support scale-to-zero) to keep the blog always reachable.
• service_type = LoadBalancer with session_affinity = ClientIP. WriteFreely is exposed on an external LoadBalancer IP and requests from the same client stick to the same pod.
• reserve_static_ip = true is load-bearing — do not flip it to false. The entrypoint bakes its resolved public host into config.ini at every boot, falling back to the foundation-injected GKE_SERVICE_URL when WF_PUBLIC_URL is unset. GKE_SERVICE_URL is only computed from the reserved static IP; with reserve_static_ip = false it can fall back to an unreachable internal *.svc.cluster.local hostname if the ephemeral LoadBalancer IP isn't known yet at apply time — confirmed on WriteFreely (see §6).
• NFS and Redis are off by default (enable_nfs = false, enable_redis = false). WriteFreely needs no shared filesystem: content lives in MySQL and its encryption keys are seeded from Secret Manager on every boot.
• No admin account is created automatically. Registration is closed; create the first account as a post-deploy step (see §3).
• WriteFreely is Go — Redis and PHP settings are inert. Turning enable_redis on only injects a REDIS_HOST WriteFreely ignores, and the php_* variables come from the module scaffold and are not consumed by WriteFreely.

---

2. Google Cloud Services & How to Explore Them

All commands assume you have run gcloud container clusters get-credentials <cluster> --region <region> --project <project> and that PROJECT, REGION, and NAMESPACE are set. The namespace and other identifiers are reported in the deployment Outputs.

A. GKE Autopilot — the WriteFreely workload

WriteFreely pods are scheduled on Autopilot, which bills for the CPU/memory the pods actually request. Horizontal Pod Autoscaling sizes the deployment between the minimum and maximum replica counts.

• Console: Kubernetes Engine → Workloads → select the WriteFreely workload to see pods, revisions, and events. Kubernetes Engine → Services & Ingress shows the external IP.
• CLI:

  kubectl get pods,svc,hpa -n "$NAMESPACE"
  kubectl logs -n "$NAMESPACE" deploy/<service-name> --tail=100
  kubectl describe hpa -n "$NAMESPACE"          # current vs target utilisation

See App_GKE for how Autopilot, scaling, and the workload type (Deployment vs StatefulSet) are managed.

B. Cloud SQL for MySQL 8.0

WriteFreely stores all application data (blogs, posts, users, sessions) in a managed Cloud SQL for MySQL 8.0 instance. Pods reach it privately through the Cloud SQL Auth Proxy sidecar over 127.0.0.1:3306; no public IP is exposed. On first deploy an initialization Job (db-init) creates the application database and user; the container entrypoint then runs writefreely db init to build the tables.

• Console: SQL → select the instance for connections, backups, flags, and metrics.
• CLI:

  gcloud sql instances list --project "$PROJECT"
  gcloud sql instances describe <instance-name> --project "$PROJECT"
  # Open an interactive shell to inspect schema/data:
  gcloud sql connect <instance-name> --user=<db-user> --database=<db-name> --project "$PROJECT"

The instance name, database, user, and the Secret Manager secret holding the password are all surfaced in the Outputs. For the connection model, automated backups, and password rotation, see App_GKE.

C. Cloud Storage

A dedicated Cloud Storage data bucket (writefreely-uploads) is provisioned automatically and the workload service account is granted access. Additional buckets can be declared via storage_buckets.

• Console: Cloud Storage → Buckets.
• CLI:

  gcloud storage buckets list --project "$PROJECT"
  gcloud storage ls gs://<data-bucket>/          # bucket name is in the Outputs

See App_GKE for CMEK options and GCS Fuse mounts.

D. Secret Manager

Three cryptographic secrets are generated automatically and stored in Secret Manager — the AES-256 keys WriteFreely uses to sign session cookies (cookies-auth), encrypt cookie payloads (cookies-enc), and encrypt stored email addresses (email-key). They are delivered to the pod via the Secret Store CSI driver and injected as WF_KEY_COOKIES_AUTH, WF_KEY_COOKIES_ENC, and WF_KEY_EMAIL. The database password is managed separately by the foundation.

• Console: Security → Secret Manager.
• CLI:

  gcloud secrets list --project "$PROJECT" \
    --filter="name~cookies-auth OR name~cookies-enc OR name~email-key"
  gcloud secrets versions access latest --secret=<secret-name> --project "$PROJECT"

See App_GKE for the Secret Store CSI integration and rotation, and WriteFreely_Common for why these keys must stay stable.

E. Networking & ingress

By default the workload is exposed through an external Cloud Load Balancing IP (service_type = LoadBalancer). A custom domain with a Google-managed certificate can be enabled, and a static IP can be reserved so the address survives redeploys.

• reserve_static_ip defaults to true — do not change it to false. WriteFreely bakes a self-referencing public host into config.ini at every container boot (the entrypoint's PUBLIC_URL fallback chain: WF_PUBLIC_URL → CLOUDRUN_SERVICE_URL → GKE_SERVICE_URL). App_GKE computes GKE_SERVICE_URL from the reserved static IP when reserve_static_ip = true; with false, the ephemeral LoadBalancer IP is often not yet known at the moment Terraform renders the Deployment's env vars, so GKE_SERVICE_URL falls back to the cluster-internal *.svc.cluster.local hostname — a real, reproducible race, not just a theoretical edge case (confirmed on WriteFreely and fixed by setting reserve_static_ip = true, per the module's deploy.tfvars default). An unreachable internal host baked into federation/generated links looks like "the app works but every absolute link is broken."
• Console: Network services → Load balancing; VPC network → IP addresses.
• CLI:

  kubectl get ingress,svc -n "$NAMESPACE"
  gcloud compute addresses list --project "$PROJECT"

See App_GKE for custom domains, Cloud CDN, and static IP details.

F. Cloud Logging & Monitoring

Pod stdout/stderr flow to Cloud Logging; GKE and Cloud SQL metrics flow to Cloud Monitoring. Optional uptime checks and alert policies are available. The entrypoint logs its progress (WriteFreely: rendered config.ini …, … seeded stable encryption keys …, … starting server …), which is useful when diagnosing first boot.

• Console: Logging → Logs Explorer; Monitoring → Dashboards / Alerting.
• CLI:

  gcloud logging read 'resource.type="k8s_container" AND resource.labels.namespace_name="'"$NAMESPACE"'"' \
    --project "$PROJECT" --limit 50

---

3. WriteFreely Application Behaviour

• First-deploy database setup. An initialization Job runs db-init.sh using mysql:8.0-debian. It creates the application database and user, grants ALL PRIVILEGES on the database, verifies the app user can connect, and shuts down the Cloud SQL Proxy sidecar. The job is idempotent (CREATE ... IF NOT EXISTS, max_retries = 3) and safe to re-run.
• Schema created on start. The container entrypoint renders config.ini (with DB_HOST = 127.0.0.1) and then runs writefreely db init on every start to create the tables (tolerant if they already exist), so the schema is bootstrapped without a separate migration step.
• The three AES-256 keys are immutable after first boot. They are generated once and written to Secret Manager. Changing any of them logs out every user (cookie signatures no longer validate) and makes previously encrypted email addresses undecryptable. Only rotate during a planned maintenance window.
• Create the first account after deploy. Registration is closed by default (open_registration = false) and no admin is seeded. To create the first account, either temporarily set WF_OPEN_REGISTRATION = "true" via environment_variables, register through the UI, then set it back to "false"; or exec WriteFreely's --create-admin inside a running pod:

  kubectl exec -n "$NAMESPACE" deploy/<service-name> -- \
    /usr/local/bin/writefreely --create-admin <user>:<password>

• Public URL correctness. The entrypoint falls back to the foundation-injected GKE_SERVICE_URL for the public host. After the LoadBalancer IP is assigned, set WF_PUBLIC_URL (via environment_variables) to the external URL or custom domain so generated links and federation use the reachable host.
• Health path. The startup probe is TCP (Ready as soon as port 8080 is bound) and the liveness probe is HTTP GET / — WriteFreely serves its home page with a 200 when healthy; there is no dedicated /health endpoint.
• Confirm the injected DB host in the running pod:

  kubectl exec -n "$NAMESPACE" deploy/<service-name> -- env | grep -E 'DB_HOST|WF_'

---

4. Configuration Variables

Variables are grouped exactly as they appear on the deployment platform. Only settings specific to or notable for WriteFreely are listed; every other input is inherited from App_GKE with its standard behaviour and defaults.

Group 3 — Application Identity

Variable	Default	Description
application_name	writefreely	Base name for resources. Do not change after first deploy.
application_version	latest	writeas/writefreely image tag; latest resolves the base image to the pinned 0.12.0 build ARG. Pin a release in production.

Group 4 — Runtime & Scaling

Variable	Default	Description
deploy_application	true	Set false to provision infrastructure only.
container_image_source	custom	Leave as custom — the config-gen wrapper must be built.
min_instance_count	1	Minimum replicas; GKE does not support scale-to-zero.
max_instance_count	1	Maximum replicas.
cpu_limit / memory_limit	1000m / 2Gi	Per-pod resources.
container_port	8080	WriteFreely's web server binds port 8080.
enable_cloudsql_volume	true	Auth Proxy sidecar; DB_HOST is overridden to 127.0.0.1.

Group 6 — GKE Backend & Cluster

Variable	Default	Description
service_type	LoadBalancer	External IP exposure for the blog.
session_affinity	ClientIP	Sticky routing to the same pod.
workload_type	null	Defaults to a stateless Deployment.

Group 5 — Environment Variables & Secrets

Variable	Default	Description
environment_variables	{}	Extra settings. Use for WF_SITE_NAME, WF_SITE_DESCRIPTION, WF_PUBLIC_URL, WF_OPEN_REGISTRATION. Do not set WF_KEY_* or DB_* here.
secret_environment_variables	{}	Map of env var → Secret Manager secret name.

Group 13 — Filesystem (NFS)

Variable	Default	Description
enable_nfs	false	Off by default: WriteFreely needs no shared filesystem. Content lives in MySQL and the encryption keys are seeded from Secret Manager on every boot.
nfs_mount_path	/var/lib/writefreely	Mount path inside the container.

Group 16 — Database Backend

Variable	Default	Description
application_database_name	writefreely	Database name → injected as DB_NAME. Immutable after first deploy.
application_database_user	writefreely	Application user → injected as DB_USER. Immutable after first deploy.
database_type	"MYSQL_8_0"	Set explicitly, matching the engine the shared application layer expects.

Group 10 — Observability & Health

Variable	Default	Description
startup_probe	TCP, 30s delay	Ready as soon as port 8080 is bound.
liveness_probe	HTTP /, 30s delay	Also drives readiness and the load-balancer health check, so a long delay keeps the pod out of service; the TCP startup probe already covers slow boots.

Group 15 — Redis (inert for WriteFreely)

Variable	Default	Description
enable_redis	false	Leave off — WriteFreely cannot use Redis; enabling it only injects an ignored REDIS_HOST.

All other inputs follow standard App_GKE behaviour.

---

5. Outputs

These values are returned on a successful deployment and are the quickest way to locate and explore the running resources.

Output	Description
service_name	Kubernetes Service name.
namespace	Namespace the workload runs in.
service_cluster_ip	In-cluster ClusterIP.
stage_service_cluster_ips	Map of ClusterIPs for stage-specific services.
service_external_ip	External LoadBalancer IP (when a static IP is reserved).
service_url	URL to reach WriteFreely.
database_instance_name	Cloud SQL instance name.
database_name	Application database name.
database_user	Application database user.
database_password_secret	Secret Manager secret holding the DB password.
database_host / database_port	DB endpoint (127.0.0.1 via the Auth Proxy) / port.
storage_buckets	Created Cloud Storage buckets.
network_name / network_exists / regions	VPC network, presence, available regions.
container_image / container_registry	Deployed image and Artifact Registry repo.
monitoring_enabled / monitoring_notification_channels	Monitoring status and channels.
initialization_jobs / db_import_job	Names of the setup and (optional) import jobs.
deployment_id / tenant_id / resource_prefix	Naming identifiers.
project_id / project_number	Project identifiers.
cicd_enabled / cicd_configuration	CI/CD status and details (repo, trigger, registry).
github_repository_url / github_repository_owner / github_repository_name	CI/CD GitHub details.
artifact_registry_repository / cloudbuild_trigger_name / cloudbuild_trigger_id	Registry and build trigger.
kubernetes_ready	Whether the cluster/workload is ready.
vpc_sc_enabled / vpc_sc_perimeter_name / vpc_sc_dry_run_mode	VPC-SC status.
audit_logging_enabled / artifact_registry_cmek_enabled	Audit logging and CMEK status.

---

6. Configuration Pitfalls & Sensible Defaults

Risk: Critical (data loss / outage / security) — High (service degraded) — Medium (cost or partial degradation) — Low (minor).

Inherited plan-time validation. This module passes its configuration through the App_GKE foundation engine, which validates values and combinations at plan time — IAP with no authorized identities, a gen1 runtime with NFS/GCS mounts, a database_type that does not match an enabled extension, memory quota values without binary unit suffixes, an out-of-range backup_retention_days. Invalid configuration fails the plan with a clear, named error before any resource is created, so most mistakes below are caught up front rather than at apply or runtime.

Setting	Sensible value	Risk	Consequence if wrong
AES-256 keys (WF_KEY_*, auto-generated)	Never rotate after first boot	Critical	Rotating logs out every user and makes encrypted email data undecryptable.
application_database_name / application_database_user	Set once	Critical	Immutable after first deploy; renaming recreates the DB/user and destroys all data.
enable_cloudsql_volume	true	Critical	On GKE the Auth Proxy sidecar is required; disabling it removes the 127.0.0.1:3306 listener and breaks MySQL connectivity.
enable_backup_import	false unless restoring	Critical	Enabling without a valid backup file fails the import job.
container_image_source	custom	High	Setting prebuilt without an image that embeds the config-gen entrypoint yields a pod that cannot render config.ini and fails to start.
WF_PUBLIC_URL	External LoadBalancer URL / domain	High	An incorrect public host breaks generated links, federation, and redirects.
reserve_static_ip	true	High	With false, GKE_SERVICE_URL can fall back to an unreachable internal *.svc.cluster.local host at apply time (an ephemeral LB IP race), which the entrypoint bakes into config.ini as the public host — confirmed on WriteFreely.
quota_memory_requests / _limits	binary units (4Gi, 8192Mi)	Critical	Bare integers are bytes and block all pod scheduling in the namespace.
min_instance_count	1	High	GKE requires min ≥ 1; the validation guard rejects invalid values.
session_affinity	ClientIP	Medium	Without stickiness, sessions may bounce between pods; stable cookie keys make this tolerable but affinity is preferred.
application_version	Pin a release	Medium	latest can shift the base image across redeploys; pinning keeps builds reproducible.
WF_OPEN_REGISTRATION	false after first admin	Medium	Leaving registration open lets anyone with the URL create an account.
enable_pod_disruption_budget	true	Medium	Disabling allows GKE to evict all pods simultaneously during maintenance.
backup_retention_days	7 (raise for prod)	Medium	Too short for compliance retention.

---

For the foundation behaviour referenced throughout — IAM and Workload Identity, autoscaling, ingress and certificates, CI/CD, Cloud Armor, IAP, Binary Authorization, VPC-SC, backups, and image mirroring — see App_GKE. WriteFreely-specific application configuration shared with the Cloud Run variant is described in WriteFreely_Common.

Related guides

• Hands-on lab: WriteFreely on GKE Autopilot — deploy it step by step, with the console screens and commands at each stage.
• WriteFreely on Google Cloud Run — the same application on Cloud Run, for when you need the other deployment target.
• WriteFreely Common — Shared Application Configuration — the configuration shared by both deployment targets.